© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 12288
    9. 8-RES_FS12288-JB
    11. 1391
    Prev
    Next
    INDEX
    Explanations

    information related to construction and cooking recipes

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    Data Type
    torch.float32
    Hook Point
    blocks.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    87
    +0.23
    1.2%
    459
    +0.22
    1.1%
    756
    +0.16
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    459
    +0.23
    0.09
    756
    +0.22
    -0.01
    87
    +0.16
    -0.03
    Negative Logits
     Freak
    -0.62
    sect
    -0.62
     Fiorina
    -0.59
     resurg
    -0.58
    ileaks
    -0.58
    Amid
    -0.57
    ghazi
    -0.57
     debunked
    -0.56
    politics
    -0.56
    cellence
    -0.56
    POSITIVE LOGITS
     optional
    0.99
     excludes
    0.99
     exceptions
    0.95
     optionally
    0.95
     varies
    0.90
     depending
    0.88
     preferably
    0.85
     omit
    0.83
     vary
    0.82
     prefer
    0.82
    Activations Density 1.334%

    No Known Activations